2335
J. gen. ViroL (1986), 67, 2335-2340. Printed in Great Britain
Key words: human rhinovirus/host range~mousemodel
Establishment of a Mouse Model for Human Rhinovirus Infection
By F A Y H . Y I N * AND N A N C Y
B. L O M A X
E. I. du Pont de Nemours & Company, Central Research & Development Department,
Experimental Station, Wilmington, Delaware 19898, U.S.A.
(Accepted 9 July 1986)
SUMMARY
W e describe here a mouse model for rhinovirus infection using a variant o f h u m a n
rhinovirus type 2 ( H R V 2 / H ) which replicated 50- to 300-fold in the lungs o f BALB/c
mice. The variant virus differed only marginally from H R V 2 / H according to various
biochemical parameters. Use of a photosensitive inoculum and p r e t r e a t m e n t of the
animals with actinomycin D were necessary for detection of reproducible and
significant levels of virus replication. This mouse model of rhinovirus infection is the
first example of h u m a n rhinovirus replication in a n o n - p r i m a t e m a m m a l , and provides
an important link for the development of rhinovirus therapy or prophylaxis.
INTRODUCTION
H u m a n rhinoviruses (HRVs), p r i m a r y causative agents of the ' c o m m o n cold' (Tyrrell &
Bynoe, 1966; Gwaltney, 1983), have a narrow host range, replicating only in higher primates. A
small animal model for H R V infection has been lacking and current in vivo tests of potential
antivirals use primates (Shipkowitz et aL, 1972) which are expensive, or mice infected with
indicator viruses (Eggers, 1976; H e r r m a n n et al., 1982) which do not cause upper respiratory
tract infections.
Because mouse cells have receptors for a few serotypes of rhinoviruses (Yin & Lomax, 1983),
we a t t e m p t e d to use the mouse as a non-primate host for rhinovirus infection. Our a p p r o a c h was
to select host range variants of h u m a n rhinovirus type 2 ( H R V 2 / H ) in tissue culture and then test
these variants for their ability to replicate in mice. Non-permissive host cells were used as the
only selective pressure; no mutagens were used. W e reported previously the successful selection
of a variant which replicates in a mouse cell line (Yin & Lomax, 1983). This p a p e r describes the
continuation of that work and the demonstration of replication of a rhinovirus variant in mice.
METHODS
Cells. HeLa cells were obtained from Flow Laboratories. They were maintained as monolayers at 37 °C in
McCoy's 5A medium (modified) (Gibco), containing 10~ heat-inactivated calf serum, gentamicin (50 ~tg/ml)and
fungisone (2.5 ~tg/ml).
L cells, strain LM, were originally obtained from the American Type Culture Collection, and have been
maintained in our laboratory since 1974. A subculture of these L cells which best supported replication of the
rhinovirus host range variant was used throughout this study. L cells were propagated as monolayers in Eagle's
MEM with Earle's salts (Gibco), supplemented with 7~ heat-inactivated foetal bovine serum, and antimicrobial
agents as listed above. Cells were karyotyped to confirm the mouse origin of this cell line.
Mice. All inbred mice were from Jackson Laboratories, Bar Harbor, Me., U.S.A. CD-I mice were from Charles
River Breeding Laboratories, Wilmington, Mass., U.S.A.
Organ culture of mouse nasal turbinates. Organ cultures were prepared by aseptically removing the turbinates
from 6-week-old, CD-1 mice and placing the turbinate tissue in culture medium without serum for 3 days before
use for virus infection.
Viruses. HRV2 strain HGP was obtained from R. Grunert (E. I. du Pont de Nemours & Co.).
Host range variants of HRV2/H were obtained as summarized in Fig. 1 and described below. Non-permissive L
ceils were infected with plaque-purified HRV2/H at an m.o.i, of 1 to 10. The starting inoculum contained about
0000-7152 © 1986 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 12:55:25
2336
F. H. YIN AND N. B. LOMAX
HRV2/H
Alternate passages between
HeLa and L cells, 40×
HRV2/L
l
Organ culture of mouse
turbinate, 2 x
HRV2/T
CD- 1 mice
3x
HRV2/m
l
Plaque purification
HRV2/M
Fig. 1. Procedure for selection of a variant HRV2/H which can replicate in the mouse.
107 p.f.u./ml HRV2/H. After one cycle of growth in L cells (24 h) the resulting virus from the L cell cytoplasm,
determined by plaque titration on HeLa cells, was about 104 p.f.u./ml. This virus was then propagated in HeLa
cells to increase the virus concentration. After 40 such alternate passages, the virus began to form very small plaques
on L cells (HRV2/L) (Yin & Lomax, 1983). L cells were used exclusively for all subsequent virus propagations.
Photosensitive HRV2/L was prepared by growing the virus in L cells in MEM, supplemented with 1~ foetal
bovine serum and neutral red (Hartman-Leddon Co., Philadelphia, Pa., U.S.A.) at 4 l.tg/ml. All operations
involving the propagation and use of photosensitive virus were carried out under red light (650 to 700 nm).
Photosensitive HRV2/L was harvested 24 h post-infection and concentrated from the L cell cytoplasm by
centrifugation (1.6 x l0 Tg-min). This virus was used to infect organ cultures of mouse turbinates. Twenty-four h
after infection, the turbinates were homogenized in 1 ml culture medium. Photosensitive virus, residual from the
inoculum, was inactivated by placing the homogenate samples in Petri dishes, with constant stirring, 8 cm from a
light source (Type BL, General Electric, 14 W, 360 nm peak) for 30 min at 4 °C. The light-insensitive progeny virus
(HRV2/T) was again propagated in L cells, made photosensitive and used to infect CD-1 mice. At 24 h postinfection, light insensitive progeny virus was recovered from the lung homogenate of infected animals. HRV2/M
was obtained by plaque isolation following passage and replication of HRV2/T in an individual CD-1 mouse, The
identities of all of the above variants were confirmed by neutralization tests with standard antisera against
HRV2/H.
Infection of mice and assay for virus growth. Each experimental group routinely consisted of six, 4-week-old
female mice. All mice were pretreated intranasally with 50 gl Cosmegen® (soluble actinomycin D; Merck, Sharpe
& Dohme) at 0.25 mg/kg body weight, at either - 24 h or - 4 h. At 0 h, all animals were infected intranasally with
50 gl photosensitive HRV2/M containing approximately 108 p.f.u. All mice were anaesthetized with ether prior to
intranasal treatment. Two mice from each group were sacrificed by cervical dislocation at 2 h after virus infection
to determine levels of residual input virus, while the remaining mice were sacrificed at 24 h. Lungs were excised
from each animal and homogenized in 3 ml culture medium and the homogenates were cleared of debris by lowspeed centrifugation.
Photosensitive virus was inactivated by exposing the homogenate to white light for 30 min; the virus titre of
each sample was then determined by plaque assay on L cells. Virus replication for each mouse was calculated as
follows. Replication index = virus titre at 24 h for individual/mean virus titre at 2 h for experimental group.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 12:55:25
Mouse model for human rhinovirus infection
2337
60
50
40
.=.
=
•~
30
~' 20
10
T
BALB/cJ
CBA/J
DBA/2J
A/J
SJL/J
BALB/
cOuJ-nude
CD-1
Fig. 2. Virus growth in various genetically inbred mice as compared with the outbred strain, CD-1.
Infection procedure was as described in Methods. All mice were pretreated with Cosmegen® at - 24 h,
with the exception of the BALB/cOuJ-nude mouse strain which was tested without Cosmegen®. Each
point on the figure represents mean replication for each group + S.E.M.
RESULTS
A rhinovirus variant with altered host range was selected in a stepwise manner using a nonpermissive host as the only selective pressure (Fig. 1). This variant was subsequently used to
infect organ cultures derived from mouse nasal turbinates. Initially, we could not detect virus
replication in the organ culture in the presence of a high background of input virus. This
difficulty was circumvented by using a photosensitive inoculum (Wilson & Cooper, 1963).
HRV2/L, containing neutral red, was found to be fully infectious when maintained in the dark,
but 99.99 ~ could be inactivated by white light. [In contrast to one published report (Madshus et
al., 1984), we found no loss of infectivity in HRV2 preparations containing neutral red if
maintained in the dark.] Thus, low levels of light-resistant progeny virus from HRV2/L-infected
turbinate organ culture could be readily distinguished from non-eclipsed light-sensitive input
virus. This same procedure employing photosensitive virus was used to obtain HRV2 able to
replicate in the mouse (HRV2/m). Finally, HRV2/M was plaque-purified and propagated in L
cells prior to use in animal experiments.
For growth and detection of virus in vivo, we used photosensitive HRV2/M as inoculum, in
mice treated intranasally with Cosmegen® at 0.25 mg/kg body weight 24 h before infection.
HRV2/M was found to replicate about 10-fold in CD-1 mice; however, we found that genetically
inbred mice supported more reproducible and vigorous virus growth. Inbred mouse strains
BALB/cJ, CBA/J, DBA/2J, A/J, SJL/J and BALB/cOuJ-nude were tested. The results are
presented in Fig. 2. Although BALB/cOuJ-nude mice supported virus replication without
Cosmegen® pretreatment, BALB/cJ mice, with Cosmegen® pretreatment, provided the best
virus yield and reproducibility and were therefore chosen as the preferred strain for routine
studies.
Higher virus replication and reduced variability were repeatedly found in Cosmegen®-treated
animals. The correlation between Cosmegen® pretreatment and virus replication is shown in
Table 1. In later experiments, we observed several-fold additional enhancement of virus
replication when Cosmegen® was administered intranasally at only 4 h prior to virus infection,
as demonstrated by the results in Table 2.
To determine the kinetics of virus replication, BALB/cJ mice were pretreated intranasally
with Cosmegen®, followed by intranasal virus administration after 4 h. As shown in Fig. 3, virus
titre in the mouse lungs reached a maximum at 24 h after intranasal infection, and then declined
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 12:55:25
2338
F. H. YIN AND N. B. LOMAX
I
106
I
I
[
~
I
~
!
lOs
&
>
10 3
I
0
6
1
I
I
l~
I
~.~
12
18
24
36
Time after infection (h)
I
•
48
72
Fig. 3. Kinetics of HRV2/M replication in BALB/cJ mice. Infection procedure was as described in
Methods, except all mice were pretreated with Cosmegen® at - 4 h. Groups of five mice were sacrificed
at various times after virus administration and lung homogenates were assayed for light-insensitive
virus. 0 , Titre from individual animals; 7q, geometric mean titre for groups of animals at specified
times.
T a b l e 1.
Correlation between Cosmegen®pretreatment and virus replication & HR V2/M-infected
mice
Mice
Cosmegen
pretreatment
BALB/cJ
None
BALB/cJ
-24 h
Virus titre*
(p.f.u./ml)
Sample
Lung
Lung
Lung
Lung
Lung
2
24
2
24
48
h
h
h
h
h
2.1
2.7
4.6
1.8
2.0
x
x
×
×
x
Virus replication
(+S.E.M.)
103
104
103
105
105
12.9 (+ 5.7)
39.0 (+11.7)
43-0 (+ 10"7)
* Virus titres are presented as the geometric mean for each group.
T a b l e 2.
Replication of HR V2 in mouse lung and nasal turbinates
Mice
Cosmegen
pretreatment
Inoculum
BALB/cJ
--4 h
HRV2/M
BALB/cJ
-4 h
HRV2/H
Sample*
Virus titret
(p.f.u./ml)
Lung 2 h
1.1
Lung 24 h
3-9
Turbinate 2 h <5.1
Turbinate 24 h
5-1
Lung 2 h
5.9
Lung 24 h
<6.3
×
x
x
x
X
x
103
l0 s
l0 t
103
102
101
Virus replication
(+S.E.M.)
355-0 (+71.4)
> 100
0
* Lung tissue and turbinate tissue of the infected mice were aseptically removed and homogenized in 3 ml and
1 ml culture medium, respectively, and virus titre determined.
1"Virus titres are presented as the geometric mean for each group.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 12:55:25
Mouse model for human rhinovirus infection
2339
slowly over the next 48 h. The variability of virus titres among animals increased greatly after
24 h. The titre reached a minimum at 8 h post-infection, but since this model was to be used for
testing antivirals (usually administered at less than 8 h), the titre at 2 h was used to calculate the
viral replication index.
It was important to demonstrate that the virus that replicated in the mouse was the selected
host range variant. This was done by infecting groups of BALB/cJ mice with either the selected
variant HRV2/M or the parental virus HRV2/H. As indicated in Table 2, virus titres of lung
homogenates from mice infected with HRV2/M increased > 300-fold during the 24 h period
whereas titres from mice infected with HRV2/H decreased about 10-fold in the same time
period. These results clearly demonstrate that only the selected host range variant is capable of
significant replication in the mouse lung. Importantly, the replication of the variant also occurs
in mouse nasal turbinates (Table 2), resembling viral replication during H R V infection.
We reported previously that the restriction of HRV2/H replication in mouse L cells occurs
during early viral replicative events subsequent to viral adsorption and penetration and that the
only measurable alteration of the host range variant HRV2/L is in a non-structural protein, P25b (Yin & Lomax, 1983). Differences between virions of HRV2/H and HRV2/L or HRV2/M
appear to be minor; we could detect no changes in antigenicity or acid lability. HRV2/M has an
increase in the apparent molecular weight of a non-structural protein (P2-5b), identical to the
alteration observed in HRV2/L; the function of the protein is as yet unknown. More detailed
comparisons of the RNAs are in progress.
DISCUSSION
W e have sclcctcd a host range variant of H R V 2 / H , which diffcrs only marginally from
H R V 2 / H by various biochcmical paramctcrs, and which can replicate 50- to 300-fold in the
mouse lung. The proccdurc for selecting host range variants is gcncrally applicable to other
rhinoviruscs for which L cells havc receptors. For example, variants of H R V I A have bccn
sclcctcd and shown to replicate more than 100-fold in the mousc lung (same procedure; F. H.
Yin & N. B. Lomax, unpublishcd results).Such a mousc model of H R V infcction should bc
useful for testing thc efficacy of antiviral compounds or potcntial vaccines.
Although the mcchanism for its effectis not well undcrstood, actinomycin D prctrcatment is
an important facet of our model. For thc tcstingof some antivirals,thispretrcatment could have
a confounding cffcct.The dosagc used has no apparent long-term toxic effectson the animals.
Cosmcgen® may act to supprcss host dcfcnces, sincc prctrcatmcnt of thc mice with
indomcthacin, a known immunosupprcssivc agent, yicldsparallelvirus replicationrcsults(F. H.
Yin & N. B. Lomax, unpublished data).
Other variablcs in our cxpcrimcnts should also bc mcntioncd. Thc host range variant,
HRV2/L, has bccn shown in tissueculture to bc somewhat lesssensitivethan thc parent strain to
thc antiviraleffectsof compounds which inhibit viral R N A synthesis (Yin & Lomax, 1983). The
diffcrcnces arc not grcat, howcvcr; the 90~o tissue culture effective doses for the two viruscs
have ncvcr bccn greatcr than twofold for any inhibitor tested. In addition, there is somc
variability of virus rcplication between cxperiments using diffcrcnt inocula (e.g. 355-fold
replication in Tablc 2 and 36-fold replication in Fig. 3). This probably reflccts the levels of
residual photoinscnsitivc virus bctwccn inocula, which would amplify the variability of thc
resultant virus replication index. In our hands, cxpcrimcnts performed using identical inocula
routinely yielded excellent reproducibility.
This mouse model is the first reported example of HRV replication in small laboratory
animals. Moreover, the model provides an acccurate and reproducible in vivo assay of HRV2
replication which is much more convenient than the currently available alternatives
(Shipkowitz et al., 1972; Eggers, 1976; Herrmann et al., 1982).
We greatly appreciate the support of the followingcolleagues from du Pont: Dr Richard Grunert and Charles
McDaniel of Biomedical Products Department for performing animal work in the early phase of this project;
Delores Tatman for excellent work with the animals; Drs Karl K. Lonberg-Holm, Bruce D. Korant and James C.
Kauer for critical review of this manuscript.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 12:55:25
2340
F.H.
YIN AND N. B. LOMAX
REFERENCES
EGGERS,H. J. (1976), Successful treatment of enterovirus-infectedmice by 2-(~-hydroxylbenzyl)-benzimidazoleand
guanidine. Journal of Experimental Medicine 143, 1367-1381.
GWALTNEY, J. M., JR (1983). Rhinovirus colds: epidemiology, clinical characteristics and transmission. European
Journal of Respiratory Disease 64 (supplement 128), 336-339.
HERRMANN, E. C., JR, HERRMANN, J. k. & DELONG, D. C. (1982). Prevention of death in mice infected with
coxsackievirus A 16 using guanidine HC1 mixed with substituted benzimidazoles. Antiviral Research 2, 339346.
MADSHUS,I. H., OLSNES,S. & SANDVIG,K. (1984). Different pH requirements for entry of the two picornaviruses,
human rhinovirus 2 and murine encephalomyocarditis virus. Virology 139, 346-357.
SHIPKOWITZ, N. L., BOWER, R. R., SCHLEICHER, J. B., AQUINO, F., APPELL, R. N. & RODERICK, W. R. (1972). Antiviral
activity of a bis-benzimidazole against experimental rhinovirus infections in chimpanzees. Applied
Microbiology 23, 117-122.
TYRRELL, D. A. J. & BYNOE,M. L. (1966). Cultivation of viruses from a high proportion of patients with colds. Lancet
i, 76-77.
WILSON,J. N. & COOPER,P. D. (1963). Aspects of the growth of poliovirus as revealed by the photodynamic effects of
neutral red and acridine orange. Virology 21, 135-145.
YIN, F. H. & LOMAX,N. B. (1983). Host range mutants of human rhinovirus in which nonstructural proteins are
altered. Journal of Virology 48, 410~,18.
(Received 20 March 1986)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 12:55:25